Many vehicle components cannot be checked after they are inserted or installed. Contamination from clinging particles can cause a total failure. Efficient cleanliness measurement for quality assurance of the manufacturing processes cannot be carried out directly on the complex part. Therefore, the degree of contamination is measured indirectly by means of microscopic residual dirt analysis. This is done in three steps:
- Rinsing/washing the parts
- Filtering the rinse liquid
- Optically analysing the filter
In close cooperation with suppliers to the automotive industry, Leica Microsystems has developed the complete Leica Cleanliness Expert system for residual dirt analysis. It consists of an automated microscope with scanning stage and digital camera and a high-performance computer with corresponding analytical software that measures the size and number of dirt particles on a filter. Leica Cleanliness Expert analyses particulate filters according to the specifications of VDA 19 and ISO/DIS 16 232. The versatile software can also be adapted quickly and easily to other industry standards or internal factory quality standards. The tester is guided through the measurement routine until the automatic particle measurement is started. The results are displayed automatically (Figure 1).
Figs. 1a–c: The results of the particle measurement are displayed automatically.a) Analysis configuration obtained by setting the threshold value in the live image. b) Composition of the full mosaic. The intelligent detection algorithm follows particles and fibers beyond the image frames. c) The result controller in the live image ensures reliable results.
The 10-pixel criterion
In accordance with ISO/DIS 16 232, in automated particle measurement, the length of the smallest particles should be rendered by at least 10 pixels (Figure 2). The corresponding calibration value for the particle size depends on the magnification level of the microscope, including the factor of the camera adapter and the pixel size of the camera. The calibration value can be determined manually or calculated from the system parameters using the software.
To calculate the calibration value, the pixel size of the camera chip is divided by the overall optical magnification level. A pixel size of 4.65 μm and a 5x magnification level with a C-mount factor of 0.63 results in a calibration value of 1.48 μm/pixel. According to the 10-pixel criterion and the resolution of 5x lens particles sizes up from 15 to 20 µm can be measured. However, for small calibration values, such as 0.3 μm/pixel, it is absolutely necessary to ensure that according to the 10-pixel criterion, particles with a size of 3 μm are correctly detected only if an optical system with corresponding resolution is used. Details that cannot be resolved optically cannot be resolved at a later stage, either.
For manual calibration, the value is determined from a defined distance measured, if possible, down to the pixel. Generally speaking, for fixed optics, the theoretical and manually determined calibration values do not differ from each other until the second decimal place. For stereomicroscopes and macroscopes, the value should be determined manually, as the difference is greater due to the zoom optics.
The 10-pixel criterion applies primarily when measuring function-critical particles. For other particle measurements, the requirement can be attenuated to five times the value of the resolution.
For measurements in small particle classes (CCC classes A–C or larger), we recommend using fully automated light microscopes with high imaging quality. Depending on the particle size, magnification levels of 5x, 10x or 20x can be used. For larger particle classes (CCC class D or higher), zoom based systems (Leica DMS300 Cleanliness Expert or Leica DMS1000 Cleanliness Expert) are also suitable.